Hi Vladimir,
This looks good, I recall we discussed a similar topic on panama-dev mailing
list [1] to give Java developers a handle to write custom target-specific
logic. The approach is identical to .NET hardware intrinsics[2] though in this
case, we deal in higher-level target-specific intrin
-bcc core-libs-dev as I don't think future discussions of this topic
needs to include that group.
I have been catching up on these discussions now that we have reached
JDK 24 RDP.
Martin,
I would like to see some of this captured in the Alternatives section in
the JEP, particularly Xuelei's
On Thu, 12 Dec 2024 22:54:59 GMT, Martin Balao wrote:
>>> In rare situations, a third-party provider can override
>>> java.security.Provider.Service::newInstance and return an unvetted service
>>> implementation (SPI).
>>
>>
>> Well, there is a concern of mine. I don't agree the case is rare.
On Thu, 12 Dec 2024 08:04:38 GMT, Sorna Sarathi wrote:
> This PR adds missing Classpath exception statements in several files.
>
> JDK Issue: [JDK-8346069](https://bugs.openjdk.org/browse/JDK-8346069)
This pull request has now been integrated.
Changeset: 09c29d1d
Author:Sorna Sarathi
Comm
Seems what we desire here is an "exact or alternative value" model instead of
an "avoid overflow" model. This sounds interesting - can you showcase any
related ruby behavior using this model?
I think an alternative can be *IfExact that returns a nullable value, if we do
not need more informatio
In some circumstances ClassFileTransformer.transform can get
ClassCircularityError or LinkageError concatenating strings (see JDK-8264667,
JDK-8262002).
VerifyLocalVariableTableOnRetransformTest fails sometimes on Oracle CI.
The fix adds handling of the errors to get information for analysis.
--
> The double check needs less time vs enter to 'synchronize' block when it not
> needed. This patch for the xml.validation workload for runs with 112 threads
> reports ~2.5% improvement on the server with Xeon 8480:
> orig: 1T - 433.64 ops/m, 112T - 17344.25 ops/m
> patched: 1T - 437.01 ops/m, 11
On Thu, Dec 12, 2024 at 6:22 PM Ethan McCue wrote:
> Placement in the standard library aside, I think Java needs value types
> and patterns before a non-throwing version of Math.multiplyExact can be
> made that achieves the same semantic goals.
>
> switch (Maths.multiplyExact(a, b)) {
> case
Placement in the standard library aside, I think Java needs value types and
patterns before a non-throwing version of Math.multiplyExact can be made
that achieves the same semantic goals.
switch (Maths.multiplyExact(a, b)) {
case Product.of(long result) -> {}
case Product.overflow() -> {}
The double check needs less time vs enter to 'synchronize' block when it not
needed. This patch for the xml.validation workload for runs with 112 threads
reports ~2.5% improvement on the server with Xeon 8480:
orig: 1T - 433.64 ops/m, 112T - 17344.25 ops/m
patched: 1T - 437.01 ops/m, 112T - 17771
Consider:
META-INF/versions/9/com/foo/Bar.class has class file version 69.0, but
class file version 53.0 or less is required to target release 9 of the
Java Platform
or, to use a term of art from the JAR File Spec:
com/foo/Bar.class in versioned directory META-INF/versions/9 has class
file
On Mon, 9 Dec 2024 10:55:13 GMT, Taizo Kurashige wrote:
>> To resolve java/io/File/GetXSpace.java failure, I fix libGetXSpace.c to use
>> Cygwin’s `df` to get the size for comparison if the test target drive is a
>> CD-ROM drive.
>>
>> As described in JDK-8343342, GetDiskSpaceInformationW can
Hey folks, I was looking into a performance bug report for JRuby that did a
lot of numeric work (a prime number detector) and ran into a bottleneck due
to our use of Math.multiplyExact.
Basically, the *Exact methods work great as long as you never actually
overflow. If you overflow (too much), the
On Thu, 12 Dec 2024 20:53:17 GMT, Xue-Lei Andrew Fan wrote:
> > For third-party providers that override java.security.Provider::getService
> > or java.security.Provider::getServices to return services that have not
> > been evaluated against the filter or are evaluated and not allowed, a
> > s
On Tue, 10 Dec 2024 19:35:21 GMT, Justin Lu wrote:
>> Please review this PR which contains the open L10n drop changes for RDP1.
>>
>> I recommend viewing the improved diffs which are built out by Jon's tool
>> here: https://cr.openjdk.org/~jlu/output/. As always, I can not confirm the
>> corr
On Thu, 12 Dec 2024 16:27:23 GMT, Christian Stein wrote:
>>> `release of {0} too high: {1}`
>>
>> I think this is better. Probably good enough.
>>
>> My concern stems from the fact that 'release' doesn't seem to be defined for
>> _class files_. I see that the term 'release' exists, and maps to
On Thu, 12 Dec 2024 08:04:38 GMT, Sorna Sarathi wrote:
> This PR adds missing Classpath exception statements in several files.
>
> JDK Issue: [JDK-8346069](https://bugs.openjdk.org/browse/JDK-8346069)
Marked as reviewed by iris (Reviewer).
-
PR Review: https://git.openjdk.org/jdk/
On Thu, 12 Dec 2024 08:04:38 GMT, Sorna Sarathi wrote:
> This PR adds missing Classpath exception statements in several files.
>
> JDK Issue: [JDK-8346069](https://bugs.openjdk.org/browse/JDK-8346069)
This looks correct to me. The missing Classpath exception was not intentional.
This should be
On Mon, 2 Dec 2024 21:05:53 GMT, Martin Balao wrote:
> getService/getServices API overrides are supported since the initial PR.
> Please check the JEP and implementation, and let us know if you see any flaw.
I guess you refer to the following section in the JEP. Otherwise, please let
me know
On Thu, 12 Dec 2024 08:04:38 GMT, Sorna Sarathi wrote:
> This PR adds missing Classpath exception statements in several files.
>
> JDK Issue: [JDK-8346069](https://bugs.openjdk.org/browse/JDK-8346069)
Marked as reviewed by asemenyuk (Reviewer).
-
PR Review: https://git.openjdk.org
On Thu, 12 Dec 2024 00:20:13 GMT, Chen Liang wrote:
>> Improve documentation for constant pool entries. This include some models
>> for those entries, symbolic descriptors, and some general guidance in
>> package summary.
>>
>> APIDiff: https://cr.openjdk.org/~liach/apidiff/cf-consts/
>> Java
On Thu, 12 Dec 2024 18:30:54 GMT, Severin Gehwolf wrote:
>> Please review this extension to #22609 which now disallows `ALL-MODULE-PATH`
>> without explicit `--module-path` option or a non-existent module path. In
>> addition, this fixes a bug mentioned in #22609 when `ALL-MODULE-PATH` and
>>
On Thu, 12 Dec 2024 17:06:58 GMT, Severin Gehwolf wrote:
>> Please review this extension to #22609 which now disallows `ALL-MODULE-PATH`
>> without explicit `--module-path` option or a non-existent module path. In
>> addition, this fixes a bug mentioned in #22609 when `ALL-MODULE-PATH` and
>>
> Please review this extension to #22609 which now disallows `ALL-MODULE-PATH`
> without explicit `--module-path` option or a non-existent module path. In
> addition, this fixes a bug mentioned in #22609 when `ALL-MODULE-PATH` and
> `--limit-modules` are used in combination. It failed earlier an
On Thu, 12 Dec 2024 14:40:17 GMT, Volker Simonis wrote:
> In the past, the Graal Compiler used `sun.misc.Unsafe` but these usages were
> recently changed to `jdk.internal.misc.Unsafe` [1]. We should therefor export
> `jdk.internal.misc` to `jdk.graal.compiler` which is an upgradeable module in
On Fri, 6 Dec 2024 15:44:39 GMT, Severin Gehwolf wrote:
> Please review this fix for JEP 493 enabled JDKs related to the
> `--limit-modules` option. The existing jlink `bindservices` tests cover this
> issue. Previously they didn't run on a JEP 493 enabled JDK, since `jmods`
> folder is missin
On Thu, 12 Dec 2024 17:18:53 GMT, Mandy Chung wrote:
>> I went with `newLimitedFinder`. Hope that's ok.
>
> This suggestion was mainly caused as this is used by the test which raises
> the question what is a limit finder. `newModuleFinder` was the original
> method name and the javadoc specifie
On Tue, 10 Dec 2024 15:31:08 GMT, Roger Riggs wrote:
> Backport this trivial fix to parsing of the system property
> "resource.bundle.debug" to jdk24.
This pull request has now been integrated.
Changeset: 3b53ed7f
Author:Roger Riggs
URL:
https://git.openjdk.org/jdk/commit/3b53ed7fb
On Thu, 12 Dec 2024 10:06:23 GMT, Severin Gehwolf wrote:
>> Please review this fix for JEP 493 enabled JDKs related to the
>> `--limit-modules` option. The existing jlink `bindservices` tests cover this
>> issue. Previously they didn't run on a JEP 493 enabled JDK, since `jmods`
>> folder is m
On Tue, 10 Dec 2024 15:31:08 GMT, Roger Riggs wrote:
> Backport this trivial fix to parsing of the system property
> "resource.bundle.debug" to jdk24.
Marked as reviewed by lancea (Reviewer).
-
PR Review: https://git.openjdk.org/jdk/pull/22664#pullrequestreview-2500333810
> Please review this extension to #22609 which now disallows `ALL-MODULE-PATH`
> without explicit `--module-path` option or a non-existent module path. In
> addition, this fixes a bug mentioned in #22609 when `ALL-MODULE-PATH` and
> `--limit-modules` are used in combination. It failed earlier an
On Thu, 12 Dec 2024 09:57:11 GMT, Severin Gehwolf wrote:
>> Personally, I find it less clear to call it `newModuleFinder`. What it does
>> is that it creates a finder based on the passed in finder that "finds" fewer
>> modules. How about `newLimitedFinder()`? But no strong feelings, so if you
On Thu, 12 Dec 2024 10:06:23 GMT, Severin Gehwolf wrote:
>> Please review this fix for JEP 493 enabled JDKs related to the
>> `--limit-modules` option. The existing jlink `bindservices` tests cover this
>> issue. Previously they didn't run on a JEP 493 enabled JDK, since `jmods`
>> folder is m
On Tue, 10 Dec 2024 23:41:44 GMT, Mandy Chung wrote:
>> Severin Gehwolf has updated the pull request incrementally with two
>> additional commits since the last revision:
>>
>> - Handle non-existent module-path with ALL-MODULE-PATH
>> - Move test, more test fixes for JEP 493 enabled builds
>
On Wed, 11 Dec 2024 19:29:13 GMT, Quan Anh Mai wrote:
>> Hi,
>>
>> This patch improves the performance of a typical `Arena::allocate` in
>> several ways:
>>
>> - Delay the creation of the NativeMemorySegmentImpl. This avoids the merge
>> of the instance with the one obtained from the call in
On Thu, 12 Dec 2024 16:00:52 GMT, Jorn Vernee wrote:
>> According to above's prior-art, it could just read `release of {0} too high:
>> {1}`
>
>> `release of {0} too high: {1}`
>
> I think this is better. Probably good enough.
>
> My concern stems from the fact that 'release' doesn't seem to b
On Thu, 12 Dec 2024 15:34:20 GMT, Christian Stein wrote:
> `release of {0} too high: {1}`
I think this is better. Probably good enough.
My concern stems from the fact that 'release' doesn't seem to be defined for
_class files_. I see that the term 'release' exists, and maps to a versioned
dir
On Tue, 10 Dec 2024 15:31:08 GMT, Roger Riggs wrote:
> Backport this trivial fix to parsing of the system property
> "resource.bundle.debug" to jdk24.
@LanceAndersen Please review
-
PR Comment: https://git.openjdk.org/jdk/pull/22664#issuecomment-2539329603
On Thu, 12 Dec 2024 15:26:20 GMT, Christian Stein wrote:
>> In this context the actual class file major version doesn't add any value to
>> communicate the issue to the user:
>> - `javac --release N` and
>> - `jar --release N` use the "Java release feature number" in their options,
>> so does t
On Thu, 12 Dec 2024 15:32:40 GMT, Christian Stein wrote:
>> Maybe this reads better? `release version of {0} too high: {1}`:
>>
>> release version of META-INF/versions/9/version/Version.class too high: 25
>> release version of META-INF/versions/10/version/Version.class too high: 25
>>
>> Note t
On Thu, 12 Dec 2024 15:29:33 GMT, Christian Stein wrote:
>> User-facing help messages read:
>>
>> javac --help
>> ...
>> --release
>> Compile for the specified Java SE release.
>> Supported releases:
>> 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23,
On Thu, 12 Dec 2024 15:23:43 GMT, Christian Stein wrote:
>> src/jdk.jartool/share/classes/sun/tools/jar/resources/jar.properties line 94:
>>
>>> 92: unexpected versioned entry {0} for release {1}
>>> 93: error.release.value.toohigh.versioned.entry=\
>>> 94: classfile release valu
On Thu, 12 Dec 2024 14:47:24 GMT, Jorn Vernee wrote:
>> Please review this change ensuring all targeted classes in a MR-JAR file
>> should target the same or a lower classfile version.
>>
>> For example, having compiled source files with `javac` 25 without using the
>> `--release` option (or w
On Thu, 14 Nov 2024 11:57:52 GMT, Christian Stein wrote:
> Please review this change ensuring all targeted classes in a MR-JAR file
> should target the same or a lower classfile version.
>
> For example, having compiled source files with `javac` 25 without using the
> `--release` option (or wi
In the past, the Graal Compiler used `sun.misc.Unsafe` but these usages were
recently changed to `jdk.internal.misc.Unsafe` [1]. We should therefor export
`jdk.internal.misc` to `jdk.graal.compiler` which is an upgradeable module in
the JDK.
[1] Remove sun.misc.Unsafe plugins
(https://github.c
On Wed, 11 Dec 2024 08:06:19 GMT, Adam Sotona wrote:
> Class-File API lost debug printing capability with removal of the
> j.l.classfile.components.ClassPrinter in JDK-8345343.
>
> This is a backport of
> https://github.com/openjdk/jdk/commit/f88c1c6ff86b8f29a71647e46136b6432bb67619,
> review
On Wed, 11 Dec 2024 08:06:19 GMT, Adam Sotona wrote:
> Class-File API lost debug printing capability with removal of the
> j.l.classfile.components.ClassPrinter in JDK-8345343.
>
> This is a backport of
> https://github.com/openjdk/jdk/commit/f88c1c6ff86b8f29a71647e46136b6432bb67619,
> review
On Fri, 22 Nov 2024 13:42:32 GMT, Jan Lahoda wrote:
> There is a new preview language feature, `requires transitive java.base;`.
> And the `java.se` module is permitted to use the feature, without being
> marked as preview (i.e. the `java.se` module participates in preview). This
> is currentl
On Thu, 14 Nov 2024 11:57:52 GMT, Christian Stein wrote:
> Please review this change ensuring all targeted classes in a MR-JAR file
> should target the same or a lower classfile version.
>
> For example, having compiled source files with `javac` 25 without using the
> `--release` option (or wi
On Tue, 10 Dec 2024 09:10:04 GMT, Galder Zamarreño wrote:
>>> Overall, looks fine.
>>>
>>> So, there will be `inline_min_max`, `inline_fp_min_max`, and
>>> `inline_long_min_max` which slightly vary. I'd prefer to see them unified.
>>> (Or, at least, enhance `inline_min_max` to cover `minL`/max
On Thu, 14 Nov 2024 11:57:52 GMT, Christian Stein wrote:
> Please review this change ensuring all targeted classes in a MR-JAR file
> should target the same or a lower classfile version.
>
> For example, having compiled source files with `javac` 25 without using the
> `--release` option (or wi
On Thu, 12 Dec 2024 08:04:38 GMT, Sorna Sarathi wrote:
> This PR adds missing Classpath exception statements in several files.
>
> JDK Issue: [JDK-8346069](https://bugs.openjdk.org/browse/JDK-8346069)
@irisclark can we get review for this change.
-
PR Comment: https://git.openjdk.
On Thu, 12 Dec 2024 08:04:38 GMT, Sorna Sarathi wrote:
> This PR adds missing Classpath exception statements in several files.
>
> JDK Issue: [JDK-8346069](https://bugs.openjdk.org/browse/JDK-8346069)
LGTM
-
Marked as reviewed by amitkumar (Committer).
PR Review: https://git.open
On Tue, 10 Dec 2024 11:16:55 GMT, Eirik Bjørsnøs wrote:
> Please review this PR which relaxes the interpretation of the system property
> `stackwalk.debug` in `java.lang.StackStreamFactory` to be case insensitive.
>
> Motivation:
> Only 5 of 83 boolean system properties in `java.base` have a ca
On Tue, 10 Dec 2024 11:16:55 GMT, Eirik Bjørsnøs wrote:
> Please review this PR which relaxes the interpretation of the system property
> `stackwalk.debug` in `java.lang.StackStreamFactory` to be case insensitive.
>
> Motivation:
> Only 5 of 83 boolean system properties in `java.base` have a ca
> This patch intrinsifies `Math.max(long, long)` and `Math.min(long, long)` in
> order to help improve vectorization performance.
>
> Currently vectorization does not kick in for loops containing either of these
> calls because of the following error:
>
>
> VLoop::check_preconditions: failed:
On Wed, 11 Dec 2024 21:15:49 GMT, Mandy Chung wrote:
>> Severin Gehwolf has updated the pull request incrementally with three
>> additional commits since the last revision:
>>
>> - Remove too strong assertion
>> - Test fixes
>> - Address comments from Mandy
>
> Looks good. Minor renaming su
On Thu, 12 Dec 2024 04:33:59 GMT, Amit Kumar wrote:
> java/lang/Thread/virtual/Starvation.java is failing on s390x with Timeout
> error as continuations support is not there for s390x.
This pull request has now been integrated.
Changeset: 68aa4d44
Author:Amit Kumar
URL:
https://git
On Thu, 12 Dec 2024 04:33:59 GMT, Amit Kumar wrote:
> java/lang/Thread/virtual/Starvation.java is failing on s390x with Timeout
> error as continuations support is not there for s390x.
Thanks Alan, Viktor for the reviews.
-
PR Comment: https://git.openjdk.org/jdk/pull/22702#issuec
> Please review this fix for JEP 493 enabled JDKs related to the
> `--limit-modules` option. The existing jlink `bindservices` tests cover this
> issue. Previously they didn't run on a JEP 493 enabled JDK, since `jmods`
> folder is missing for them.
>
> The gist of the issue is that multiple `M
On Thu, 12 Dec 2024 09:23:32 GMT, Severin Gehwolf wrote:
>> src/jdk.jlink/share/classes/jdk/tools/jlink/internal/JlinkTask.java line 500:
>>
>>> 498: * {@code roots} set.
>>> 499: */
>>> 500: public static ModuleFinder limitFinder(ModuleFinder finder,
>>
>> Inlining `limitFinder`
On Tue, 10 Dec 2024 21:41:18 GMT, Tim Prinzing wrote:
>> Adds a JFR event for socket connect operations.
>>
>> Existing tests TestSocketEvents and TestSocketChannelEvents modified to also
>> check for connect events.
>
> Tim Prinzing has updated the pull request incrementally with one additiona
On Thu, 12 Dec 2024 04:33:59 GMT, Amit Kumar wrote:
> java/lang/Thread/virtual/Starvation.java is failing on s390x with Timeout
> error as continuations support is not there for s390x.
Nice catch, thank you!
-
Marked as reviewed by vklang (Reviewer).
PR Review: https://git.openjd
On Thu, 12 Dec 2024 09:47:30 GMT, Amit Kumar wrote:
> Can we consider this trivial ? It's not hotspot change as well.
This change only requires 1 reviewer, so you are good to integrate.
-
PR Comment: https://git.openjdk.org/jdk/pull/22702#issuecomment-2538404592
On Tue, 10 Dec 2024 21:41:18 GMT, Tim Prinzing wrote:
>> Adds a JFR event for socket connect operations.
>>
>> Existing tests TestSocketEvents and TestSocketChannelEvents modified to also
>> check for connect events.
>
> Tim Prinzing has updated the pull request incrementally with one additiona
On Thu, 12 Dec 2024 04:33:59 GMT, Amit Kumar wrote:
> java/lang/Thread/virtual/Starvation.java is failing on s390x with Timeout
> error as continuations support is not there for s390x.
Can we consider this trivial ? It's not hotspot change as well.
-
PR Comment: https://git.openjd
On Thu, 12 Dec 2024 05:39:04 GMT, Julian Waters wrote:
>> You need to split this up into multiple parts. One part is about removing
>> dead code. Do not comment it out, just remove it. Open a new JBS issue on
>> client-libs for removal of dead code. This should be trivial to get pushed.
>>
>>
On Thu, 12 Dec 2024 04:33:59 GMT, Amit Kumar wrote:
> java/lang/Thread/virtual/Starvation.java is failing on s390x with Timeout
> error as continuations support is not there for s390x.
Marked as reviewed by alanb (Reviewer).
-
PR Review: https://git.openjdk.org/jdk/pull/22702#pull
On Wed, 11 Dec 2024 21:11:39 GMT, Mandy Chung wrote:
>> Severin Gehwolf has updated the pull request incrementally with three
>> additional commits since the last revision:
>>
>> - Remove too strong assertion
>> - Test fixes
>> - Address comments from Mandy
>
> src/jdk.jlink/share/classes/jd
Hi all,
Trivial fix to problemlist
vmTestbase/vm/mlvm/indy/func/jvmti/mergeCP_indy2manySame_a/TestDescription.java
in virtual thread mode to make less test noisy, it's supplement for PR
https://github.com/openjdk/jdk/pull/21417
-
Commit messages:
- 8346016: Problemlist vm/mlvm/ind
On Thu, 12 Dec 2024 09:08:14 GMT, Alan Bateman wrote:
> This looks okay. Would it be possible to change the PR title to match the JBS
> issue as this issue isn't specific to s390x.
Done.
-
PR Comment: https://git.openjdk.org/jdk/pull/22702#issuecomment-2538335448
On Thu, 12 Dec 2024 04:33:59 GMT, Amit Kumar wrote:
> java/lang/Thread/virtual/Starvation.java is failing on s390x with Timeout
> error as continuations support is not there for s390x.
This looks okay. Would it be possible to change the PR title to match the JBS
issue as this issue isn't speci
Hi Kevin,
I really, really appreciate your work here.
Thanks for all the "light" you bring on the Java channels.
Davide
On 2024-12-11 21:31, Kevin Rushforth wrote:
Yes, jpackage will support either v4 or v5 as of JDK 24:
https://bugs.openjdk.org/browse/JDK-8319457
You can also see this in
On Thu, 14 Nov 2024 11:57:52 GMT, Christian Stein wrote:
> Please review this change ensuring all targeted classes in a MR-JAR file
> should target the same or a lower classfile version.
>
> For example, having compiled source files with `javac` 25 without using the
> `--release` option (or wi
This PR adds missing Classpath exception statements in several files.
JDK Issue: [JDK-8346069](https://bugs.openjdk.org/browse/JDK-8346069)
-
Commit messages:
- Added missing CPE statements
Changes: https://git.openjdk.org/jdk/pull/22704/files
Webrev: https://webrevs.openjdk.org/
75 matches
Mail list logo
